© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 9811
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to software development and programming frameworks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    168
    +0.12
    0.7%
    76
    +0.10
    0.6%
    275
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    129
    +0.12
    0.01
    405
    +0.10
    0.03
    29
    +0.10
    0.01
    Negative Logits
     Kepler
    -2.15
     Hb
    -2.03
     deaf
    -1.92
     RX
    -1.71
     Cobb
    -1.70
     ESR
    -1.68
     Osw
    -1.68
     RV
    -1.67
     conduction
    -1.65
     Isaac
    -1.65
    POSITIVE LOGITS
     Django
    4.25
     django
    4.19
    django
    3.86
    jango
    3.14
     {%
    2.47
    dj
    2.21
     admin
    2.07
    kwargs
    2.05
     quer
    2.01
     flask
    1.99
    Activations Density 0.311%

    No Known Activations